Humans' Bonding with their Companion Dogs: Cardiovascular Benefits during and after Stress

 
CONTINUE READING
Humans’ Bonding with their Companion Dogs:
Cardiovascular Benefits during and after Stress
                          Rebecca A. Campo

                            Bert N. Uchino
                 Department of Psychology and
                  Health Psychology Program
                      University of Utah

    This study examined whether having one’s companion dog pres-
    ent during and after stress posed similar cardiovascular benefits as
    having a close friend present, even when the relationship quality
    for both the companion dog and friend was highly positive. Positive
    aspects of relationship quality for participants’ dog and friend were
    not associated with one another, suggesting that these relation-
    ships exist independently. Additionally, compared to participants
    with a close friend present, those with their dog present had lower
    heart rate and diastolic blood pressure (p’s < .05) while undergo-
    ing the stressors, and tended to have lower heart rate and systolic
    blood pressure (p’s < .09) when recovering from stressors. This
    study indicates that even when relationship quality is similarly
    high for companion dogs and friends, dogs may be associated with
    greater reductions in owners’ cardiovascular reactivity to stress,
    particularly if there is a potential for evaluation apprehension in
    the human friendships. These findings support the value of the hu-
    man-companion animal relationship in promoting human welfare.

    Key words: bonding, companion dogs, cardiovascular health, stress

    Repetitive, exaggerated cardiovascular reactivity to psy-
chological stress may influence the development and progres-
sion of cardiovascular disease, and more generally, lead to
pathophysiological consequences such as metabolic changes,
increased inflammation, and immunosuppression (Player,
King, Mainous, & Geesey, 2007; Rosengren et al., 2004; Treiber
et al., 2003). Research indicates that human social support
may buffer cardiovascular responses to stress (Cohen & Wills,
Journal of Sociology & Social Welfare, December 2013, Volume XL, Number 4

                                    237
238			                       Journal of Sociology & Social Welfare
1985; Uchino, Cacioppo, & Kiecolt-Glaser, 1996) by changing
psychological processes (i.e., stress appraisals, emotions) that
enable one to cope more efficiently (Blascovich & Mendes,
2000; Lazarus & Folkman, 1984). However, social support’s
effectiveness to do so may depend on the quality of the rela-
tionship (i.e., positive, negative, or ambivalent quality) since
not all close relationships are purely positive (Campo et al.,
2009; Uchino, Holt-Lunstad, Uno, & Flinders, 2001). Likewise,
research has shown that support from individuals with whom
we have a positive relationship quality, compared to those
with an ambivalent relationship quality (i.e., consists of posi-
tive and negative aspects, more prevalent than purely negative
relationship quality), is associated with the lowest levels of
cardiovascular reactivity during stress (Holt-Lunstad, Uchino,
Smith, & Hicks, 2007; Uno, Uchino, & Smith, 2002).
    Like human social support, human–companion animal re-
lationships are associated with health benefits that exist after
controlling for physical exercise (Serpell, 1991). Additionally,
individuals with companion animals have reduced cardiovas-
cular responses to stress compared to those without compan-
ion animals (Allen, Blascovich, & Mendes, 2002; Allen, Shykoff,
& Izzo, 2001). Other research has shown that interacting with
one’s companion dog is associated with beneficial neuroendo-
crine changes in individuals, such as increases in dopamine,
oxytocin, and B-endorphin, and decreases in cortisol, a stress
hormone (Odendaal, 2000). Importantly, research has found
no difference between owners and non-owners in terms of
tobacco use, body mass index, or social economic status (i.e.,
income or education) that may explain such benefits.
    Similar to humans’ positive relationship quality, the attach-
ment felt with one’s companion animal may be a driving com-
ponent behind many of the psychological and health benefits
seen with human–companion animal relationships, including
reduced cardiovascular responses to stress. Attachment with
one’s companion animal is associated with lower rates of de-
pression and anxiety, and higher rates of happiness and self-
esteem (Crawford, Worsham, & Swinehart, 2006). However,
few studies have examined owners’ attachment or relation-
ship quality to their companion animals as a mechanism of
the physical health benefits (Krause-Parello, 2008; Nagaswa,
Mogi, & Kikusui, 2009; Winefield, Black, & Chur-Hansen,
Cardiovascular Benefits of Companion Dogs                     239
2008). Research has shown that attachment behavior (e.g.,
animal-initiated gazing) and high satisfaction with one’s com-
panion animal is associated with owners’ increased oxytocin
levels compared to owners who did not have similar associa-
tions with their companion animals (Nagaswa et al., 2009).
This is important in human health because oxytocin may be
a mechanism for the stress buffering effects of social support
(Heinrichs, Baumgartner, Kirschbaum, & Ehlert, 2003).
    It should be noted that the benefits derived from human–
companion animal relationships are not limited to the psycho-
logical and physical health outcomes between an individual
and his or her companion animal. Rather, companion animals
have also made significant contributions to aiding in social
welfare issues. In hospital settings, therapy dog visits can
help alleviate pain and distress in chronic pain patients and
increase well-being in accompanying family members (Marcus
et al., 2012). In psychiatric settings, animal-assisted therapy
(AAT) has been successfully used in patients struggling with
depression, loneliness, addiction, schizophrenia, and phobias
(Dimitrijevic, 2009). AAT can also help individuals who have
difficulties with human relationships become more respon-
sive during therapy sessions. In family therapeutic settings,
inquiring about the family’s pet can ease tension and provide
an opening to more difficult conversation topics (Dimitrijevic,
2009; Walsh, 2009). Furthermore, there are a variety of ani-
mal-assisted activity (AAA) programs that exist for improv-
ing children’s reading and communication skills, decreasing
loneliness in assisted living facilities, increasing motivation in
physical rehabilitation sessions, and increasing empathy and
prosocial behaviors in children with severe conduct disorders
(Walsh, 2009).
    The main purpose of this study was to compare the car-
diovascular benefits of having one’s companion animal or
close friend present during and while recovering from stress-
ors. Furthermore, considering the robust finding that posi-
tive human relationship quality is associated with cardio-
vascular benefits during stress, we wanted to compare these
effects specifically for individuals who were attached to their
companion animal and had a close friend of positive relation-
ship quality. We predicted that cardiovascular responses would
be at least equivalent when comparing participants who had
240			                        Journal of Sociology & Social Welfare
their companion animal present to those who had their close
friend present.

                           Methods
Participants
    A sample of individuals with companion dogs (N = 162)
were recruited from the community and the University of
Utah to participate in the study. The study was advertised in
Utah mainstream and alternative newspapers, and flyers were
posted throughout the Salt Lake City vicinity. Participants were
excluded if they had pre-existing hypertension or cardiovascu-
lar disease, used cardiovascular prescription medications, or
had a Body Mass Index > 35. Inclusion criteria were that the
participant had had their companion dog and same-sex best
friend for at least 2 years in order to ensure that the relation-
ships were not new. Participation was limited to dogs because
past research has shown that, after controlling for physi-
ological and psychosocial variables, dogs, compared to other
types of companion animals, made a significant contribution
to 1-year survival of patients who had been hospitalized for
myocardial infarction (Friedmann & Thomas, 1995). This does
not mean that other types of companion animals are not as-
sociated with health benefits (Allen et al., 2001; Castelli, Hart,
& Zasloff, 2001; McConnell, Brown, Shoda, Stayton, & Martin,
2011); rather, we limited our sample to dogs because it helped
us minimize any differences due to characteristics associat-
ed with different types of companion animals. Additionally,
using only owners as participants, instead of including non-
owners, may have limited the study’s generalizability, but the
intent was to compare the benefits of dogs with close friends,
without confounding existing differences between owners
and non-owners. The study was approved by the University’s
Institutional Review Board and all participants gave informed
consent.

Study Design
    This was a 3 (Support Condition: Dog, Friend, Alone) X
2 (Stressor Type: Active- or Passive-Coping Task) factorial
study design. The support condition was a between-partici-
pants factor, to which participants were randomly assigned.
The stressor was a within-participants factor, with the order of
Cardiovascular Benefits of Companion Dogs                     241
occurrence counterbalanced. A priori power calculations indi-
cated that a sample size of 165 would be sufficient for moder-
ate effect size (r = .30), with power = .82 for between-subjects
effects, power = .89 for within-subjects effects, and power = .82
for between-within interactions at 5% significance level.

Psychosocial Measures
     Relationship quality. Prior to randomization to the support
condition, participants’ relationship quality (see dog and
human relationship quality measures below) with their dog and
same-sex best friend was assessed to ensure that neither con-
sisted of negative or ambivalent relationship quality (Uchino
et al., 2001). No participants needed to be excluded due to their
relationship quality. The Companion Animal Bonding Scale
(CABS) and the Pet Attitude Scale (PAS) assessed the relation-
ship quality between the participant and his or her companion
dog. The CABS (Poresky, Hendrix, Mosier, & Samuelson, 1987)
measures self-reported behavior that is indicative of the bond
an owner has with his or her companion animal and the PAS
(Templer, Salter, Dickey, Baldwin, & Veleber, 1981) measures
the favorableness of attitudes towards companion animals.
Chronbach alphas for the CABS ranged from 0.76 to 0.82 in
two study samples of adults and parents and 0.93 for the PAS
in an undergraduate sample (Poresky et al., 1987; Templer et
al., 1981). Construct validity was confirmed in Poresky (1987)
by correlating the CABS with the PAS scale (r = 0.31, p < .001),
suggesting these measures assess similar, but not redundant
aspects of attitudes toward pets. Additionally, since a validat-
ed questionnaire does not exist that allows one to simultane-
ously assess both human and companion animal relationship
quality, we adapted the Social Relationship Inventory (SRI)
(Campo et al., 2009) by only focusing on emotional support,
instead of including other types of support functions (e.g., in-
strumental support). The SRI assessed how important, helpful,
and upsetting the dog and friend are when the participant
needs emotional support (i.e., provides emotional comfort,
relieves stress, or uplifts one’s mood). In an undergraduate
sample, reported Chronbach alphas were .69 and .80 for the
positivity and negativity subscales, and it was correlated with
the support (r = 0.76, p < .001) and conflict (r = 0.50, p < .001)
subscales of the Quality of Relationship Inventory (Pierce,
Sarason, Sarason, Solky-Butzel, & Nagle, 1997).
242			                        Journal of Sociology & Social Welfare
    Psychological variables. Participants’ psychological experi-
ence was captured with variables that research has indicated are
relevant to stress responses (Cacioppo & Petty, 1986; Lazarus &
Folkman, 1984), such as emotions, perceived threat and coping
appraisals, and evaluation apprehension. Emotional responses
were assessed with a measure that conceptualizes positive and
negative affect as independent dimensions (i.e., high negative
affect does not imply low positive affect) according to guide-
lines of Barrett and Russell (1998). Specifically, affect is viewed
as an interaction of activation (activated or deactivated) and
valence (pleasant or unpleasant). This results in four categories
of affect: pleasant-activated (determined, attentive, strong),
unpleasant-activated (distressed, nervous, jittery), pleasant-
deactivated (calm, at ease, relaxed), and unpleasant-deacti-
vated (bored, tired, sluggish). This measure was completed
after baseline, each stressor, and the recovery periods. (I don't
understand this sentence.) Perceived threat of the stressor
and appraisal of coping ability (Feldman, Cohen, Hamrick, &
Lepore, 2004) were assessed prior to each stressor. Evaluation
apprehension (i.e., feeling threatened, disturbed, evaluated by
the presence of one’s companion dog or friend) (Guerin, 1989)
was completed after each stressor (i.e., the alone condition did
not complete this measure).

Stressors
    The stressors were active-coping and passive-coping tasks
that are standard laboratory challenges used in psychophysi-
ological research (Sherwood, 1993). Active-coping tasks simu-
late types of stressors over which an individual has the ability
to mentally or physically influence the outcome (e.g., prepare
oneself for an upcoming job interview). Passive-coping tasks
simulate types of stressors over which an individual has no
control of the outcome (e.g., watching a loved one deliver a
bad speech). The active-coping stressor consisted of a 5-minute
mental arithmetic task in which the participant was asked to
subtract out loud by 7’s starting with a three-digit number
(e.g., starting with 732, subtract by 7’s). The participant was
instructed that the goal was to get to zero as quickly as pos-
sible, but without making any mistakes or the experimenter
would verbally alert him/her for every mistake made. The
Cardiovascular Benefits of Companion Dogs                    243
passive-coping stressor was a cold pressor task that consisted
of the participant holding a frozen ice pack to his or her fore-
head for 2 minutes. Participants were told not to remove the
ice pack until the experimenter informed him/her that it could
be removed.

Cardiovascular Measures
    Systolic blood pressure (SBP), diastolic blood pressure
(DBP), and heart rate (HR) were measured continuously during
the baseline, the stressors, and the recovery periods with a
Dinamap Model 8100 monitor (Critikon Corporation, Tampa,
FL). The Dinamap uses the oscillometric method to estimate
blood pressure (Epstein, Huffnagle, & Bartkowski, 1991).
Means of SBP, DBP, and HR for each period were averaged
across minutes to increase the reliability of these assessments.

                            Procedure
    All procedures were conducted in the participants’ homes
due to University regulations that only service animals were
allowed on campus and to help ensure that the compan-
ion dogs were more at ease than they would have been in a
university laboratory. This helped minimize issues related to
atypical or negative dog behavior that might have distracted
participants, and it helped increase the ecological validity of
the findings. Participants were randomly assigned to one of
the support conditions for the entire study: (a) companion dog
present, (b) close friend present, or (c) alone during the study.
The close friend was the same-sex best friend that the partic-
ipants had previously rated on relationship quality. Prior to
the study session, participants were asked to identify a room
in their home that was free from potential distractions (i.e.,
phone, TV, other people). If the participant was assigned to the
alone or friend condition, then his/her companion dog was
kept out of the room or the home. This did not adversely affect
the experiment, since none of the dogs that were put in a sepa-
rate room or outside of the home reacted negatively.
    Upon beginning the study, a blood pressure cuff was placed
on the participant, and after a 10-minute baseline assessment
of cardiovascular responses, the source of support (if assigned
244			                      Journal of Sociology & Social Welfare
to dog or friend condition) was introduced into the room. In
the dog condition, the participant was informed that he or she
may pet his/her dog if desired, taking care not to move exces-
sively due to the blood pressure readings. In the friend condi-
tion, the friend sat close by and was informed that he or she
could provide supportive behaviors if desired, but was asked
not to provide answers to the arithmetic task. This helped to
maintain equivalency between the dog condition and the friend
condition. Participants in the alone condition remained alone
throughout the study. The order of the consecutive stressors
(i.e., math task, cold pressor) was counter-balanced across the
study support conditions (dog, friend, alone) and each stressor
was followed by an 8-minute recovery period in which the dog
or friend remained present.

                           Results
Participant Demographics
    Data were collected on 162 participants; 3 participants
were excluded for failure to follow experiment instructions,
resulting in a final N of 159 for analyses. The majority of our
participants were Caucasian (89%) and female (75.5%), with an
average age of 30 years old, and a median household income
bracket of $35,000 to $55,000. The average length of partici-
pants’ relationship with companion dog and friend was 5.6
years and 9.5 years, respectively.

Math Performance
    A one-way ANOVA did not reveal any differences among
the support conditions for the percentage of correct subtrac-
tions nor number of attempted subtractions (all p’s >1). This
suggests that any potential differences in psychosocial or
physiological responses were not due to group differences in
effort or performance.
Psychosocial Outcomes
    Relationship quality. Participants’ scores on the CABS and
PAS indicated high levels of bonding and favorable attitudes
towards their companion dogs (Table 1). Additionally, the SRI
indicated high positive relationship quality with both the dog
and friend. Participants felt that their dogs and friends were
helpful, not upsetting, and important when needing emotional
Cardiovascular Benefits of Companion Dogs                                              245
support. A comparison of the SRI ratings for dogs and friends
revealed that dogs were rated as more important than friends
when needing emotional support (t (154) = -2.89, p = .004);
however, there were no differences between dogs and friends
on being helpful or not upsetting during emotional support.
Table 1. Correlations and Descriptive Statistics for Relationship
Quality Measures
 Measures                       1        2         3         4         5        6           7       8
 CABSa                          -
 PASb                        .68***      -
 SRI Dog-Importantc          .54***    .68***      -
 SRI Dog-Upsettingc           -.16     -.28**   -.25**       -
 SRI Dog-Helpfulc            .56***    .69***   .56***    -.32***      -
 SRI Friend-Importantc        -.03      -.09      .05      .17*      -.05       -
 SRI Friend-Upsettingc         .10      .08       .07      .20*       .03      .00           -
 SRI Friend-Helpfulc          -.03      -.07      .05       .05      -.06     .48***       -.18*    -
 M                            4.04      4.33    5.59**     1.88      4.83     5.36**       1.81    4.80
 SD                           .74       .43      .71       .92       1.13      .70         .98     .87
 Scale Range                  1-5       1-5      1-6       1-6       1-6       1-6         1-6     1-6

Note. M = means, SD = standard deviation. aCompanion Animal Bonding Scale
(Poresky et al., 1987). bPet Attitude Scale (Templer et al., 1981). cSocial Relationship
Inventory (Campo et al., 2009). Listwise N = 148. *p < .05. **p < .01. ***p < .001.

    Bivariate correlations were conducted to examine the re-
lationship among the different relationship quality measures
(Table 1). As expected, positive aspects of the SRI for one’s
dog (i.e., importance & helpfulness during emotional support)
were positively correlated with the PAS and CABS, and nega-
tively correlated with SRI dog-upsetting. Notably, the SRI pos-
itive ratings for friends were not significantly correlated with
the positive relationship aspects for dogs. This suggests that
individuals can have positive relationship quality with their
dogs, independently of their friend’s relationship quality. In
other words, it is not likely that individuals who bond with
their dogs do so because they have difficulty with their human
relationships or vice versa.
    Unless specified, the following psychological variables
were analyzed with 3 ( Dog, Friend, Alone) X 2 (Stressor: Math
Task, Cold Pressor) Analysis of Covariance (ANCOVAs) and
controlled for baseline values. Descriptive statistics for the
psychological variables by stressor type are reported in Table
246			                         Journal of Sociology & Social Welfare
2.
     Pretask coping appraisals. To analyze the pretask appraisals
of stress and coping resources, we calculated an appraisal ratio
following the guidelines of Feldman et al. (2004). An ANOVA
revealed a main effect for stressor type, F (1, 152) = 57.95, p <
.0001, 2 = .28. Overall, participants appraised the math task
as more threatening (M = .63, SE = .04) given their coping re-
sources than the cold pressor (M = .37, SE = .02). However,
no significant effects involving the support conditions were
found. Thus, all participants viewed the math task as more
stressful than the cold pressor.
     Emotional responses. Analyses were conducted on changes
(i.e., Task - Baseline) in emotional responses (i.e., unpleasant-
deactivation, unpleasant-activation, pleasant-deactivation,
pleasant-activation) to the stressors and recovery from the
stressors. For emotional responses to the stressors, there were
no significant effects for stressor type or support condition.
However, for recovery from the stressors, there was a signifi-
cant stressor main effect for unpleasant-activation (F (1, 150)
= 3.78, p = .05, 2 = .03). Specifically, participants had a larger
decrease in unpleasant-activation (i.e., feeling distressed,
nervous, and jittery) during recovery from the cold pressor (M
= -.26, SE = .03) than during recovery from the math task (M
= -.17, SE = .03). Additionally, there was a significant support
condition main effect (F (2, 150) = 3.35, p = .04, 2 = .04). Follow-
up comparisons revealed that those with their dog present had
a larger decrease in unpleasant-activation during recovery (M
= -.30 SE = .04) than those with their friend present (M = -.19,
SE = .05; p = .07) or who were alone (M = -.15, SE = .05; p = .02).
There was no significant interaction between type of stressor
and support-member condition.
     Evaluation apprehension. Next, the evaluation apprehen-
sion felt with the presence of one’s dog or friend during the
stressors was analyzed. The ‘alone’ participants did not com-
plete this measure, as there was no one present to potentially
increase evaluation apprehension (i.e., the experimenter was
out of participants’ view). A significant main effect for stress-
or type indicated that participants felt more evaluated (F (1,
104) = 24.99, p < .0001, 2 = .19), disturbed (F (1, 105) = 12.27,
p = .001, 2 = .11), and threatened (F (1, 103) = 7.29, p = .008,
2 = .07) during the math task than during the cold pressor.
Cardiovascular Benefits of Companion Dogs                        247
Additionally, a significant main effect for support condition in-
dicated that participants with their dog present felt less evalu-
ated (F (1, 104) = 17.05, p < .0001, 2 = .14) and less threatened
(F (1, 103) = 5.14, p = .03, 2 = .05) than those with a friend
present.
     These main effects were qualified by a significant Support
Condition X Stressor Type interactions for feeling evaluated (F
(1, 104) = 37.22, p < .0001, 2 = .26), disturbed (F (1, 105) = 5.39,
p = .02, 2 = .05), and threatened (F (1, 103) = 7.29, p = .008, 2 =
.07). Specifically, during the math task, participants with their
dog present felt less evaluated (p < .0001) and threatened (p =
.01) than those with their friend present (Table 2). The interac-
tion for feeling disturbed revealed that participants with their
friend present felt more disturbed during the math task than
during the cold pressor (p < .0001). These results indicate that
during the most stressful task (math), participants with their
dog present felt less evaluated about their ability to handle the
stressor than those with a friend present did.
     In summary of the psychological outcomes, the support
conditions did not alter participants' emotional responses
while undergoing the stressor. However, during recovery
from the stressors, participants with their dog present had a
larger decrease in unpleasant-activation affect (i.e., distressed,
nervous, jittery) than those with their friend present or who
were alone. Additionally, we found that participants with a
friend present felt more evaluation apprehension compared to
those with their dog present, particularly during the math task,
which was reported to be more stressful than the cold pressor.

Cardiovascular Responses
    ANCOVAs (Support Condition X Stressor Type) were also
used to analyze cardiovascular responses (SBP, DBP, & HR)
during the stressors and the recovery periods following the
stressors. For this purpose, change scores were computed (i.e.,
Task - Baseline) and the baseline average was included as a co-
variate. Descriptive statistics for the cardiovascular variables
by stressor type are reported in Table 2. Cardiovascular reactivity
during stressors. Analyses of cardiovascular reactivity during
the stressors revealed significant main effects for type of stress-
or for DBP (F (1, 150) = 3.93, p = .05, 2 = .03) and HR (F (1,
150) = 14.32, p < .0001, 2 = .09). In both cases, cardiovascular
248			                                 Journal of Sociology & Social Welfare
reactions were higher to the math task (DBP M = 6.24, SE = .52;
HR M = 7.72, SE = .51) than to the cold pressor (DBP M = 3.55,
SE = .42; HR M = .92, SE = .39). Furthermore, significant main
effects for support condition were found for DBP (F (2, 150) =
6.51, p = .002, 2 = .08) and HR (F (2, 150) = 4.70, p = .01, 2 = .06).

Table 2. Means of Psychological and Cardiovascular Variables
 Variables                              Alone           Dog            Friend
 Math Task Reactivity
   Appraisal Ratio                     .63 (.06)       .62 (.05)       .63 (.06)
    Threat                                 --        1.04a (.06)     1.27b (.07)
    Evaluation                             --        1.51a (.12)     2.78b (.13)
    Disturbed                              --         1.43 (.12)      1.74 (.13)
    Unpleasant-Deactivation           -.41 (.09)      -.50 (.08)      -.41 (.09)
    Unpleasant-Activation              .46 (.12)       .40 (.11)       .54 (.12)
    Pleasant-Deactivation             -.87 (.14)      -.99 (.13)     -1.13 (.14)
    Pleasant-Activation                .31 (.10)       .36 (.09)       .36 (.10)
    SBP ∆                            7.01 (1.39)     9.54 (1.24)     9.23 (1.37)
    DBP ∆                            3.62a (.95)     6.23b (.84)     8.86c (.93)
    HR ∆                             6.81 (.92)       6.35 (.82)     10.02 (.91)
 Cold Pressor Reactivity
   Appraisal Ratio                     .38 (.03)       .38 (.03)       .37 (.03)
    Threat                                 --         1.04 (.02)      1.02 (.02)
    Evaluation                             --         1.61 (.13)      1.71 (.14)
    Disturbed                              --         1.33 (.09)      1.22 (.10)
    Unpleasant-Deactivation           -.35 (.09)      -.43 (.08)      -.50 (.09)
    Unpleasant-Activation              .18 (.11)       .08 (.09)       .11 (.10)
    Pleasant-Deactivation             -.85 (.15)      -.67 (.14)      -.71 (.15)
    Pleasant-Activation                .11 (.11)      .27 (.10)        .23 (.11)
    SBP ∆                            4.11 (1.09)      5.73 (.97)     4.01 (1.07)
    DBP ∆                            2.62 (.76)       3.51 (.68)      4.52 (.75)
    HR ∆                              .27 (.70)         .69 (.62)     1.81 (.69)
 Math Task Recovery
    Unpleasant-Deactivation           -.16 (.09)      -.35 (.08)      -.17 (.09)
    Unpleasant-Activation             -.09 (.06)      -.29 (.05)      -.12 (.06)
    Pleasant-Deactivation             -.21 (.11)      -.08 (.10)      -.23 (.11)
    Pleasant-Activation               -.07 (.09)      -.02 (.08)      -.17 (.09)
    SBP ∆                             -.58 (.80)       .16 (.72)      2.05 (.82)
    DBP ∆                             -.96 (.70)       .04 (.59)       .69 (.67)
    HR ∆                             -1.34 (.68)      -.55 (.61)      1.07 (.70)
 Cold Pressor Recovery
    Unpleasant-Deactivation            -.08 (.09)     -.19 (.09)       .06 (.09)
   Unpleasant-Activation               -.21 (.06)     -.30 (.05)      -.26 (.05)
   Pleasant-Deactivation                .07 (.11)      .09 (.11)      -.08 (.11)
   Pleasant-Activation                 -.04 (.09)     -.08 (.09)      -.15 (.10)
   SBP ∆                                .63 (.89)      .94 (.80)       .59 (.90)
   DBP ∆                               -.19 (.63)     .52 (.57)       -.25 (.64)
   HR ∆                               -1.11 (.69)      .33 (.62)      -.06 (.71)

Note. The four affect variables and cardiovascular variables are changes scores
(Task – Baseline). Values in parentheses represent standard errors. Dashes indicate
variable was not measured for the Alone condition. SBP=systolic blood pressure,
DBP=diastolic blood pressure, HR=heart rate. Means in the same row that do not
share subscripts differ at p < .05. N=159; Alone n=49, Dog n=61, Friend n=49.
Cardiovascular Benefits of Companion Dogs                      249
Participants with their dog present had lower DBP and HR re-
activity (Dog: DBP M = 4.87, SE = .63; HR M = 3.52, SE = .58)
than those with their friend present (Friend: DBP M = 6.69, SE
= .69; HR M = 5.91, SE = .64; p < .05). The dog condition and
alone condition (Alone: DBP M = 3.12, SE = .71; HR M = 3.54,
SE = .65) did not differ from one another, although the friend
condition had higher DBP and HR reactivity than the alone
condition (p < .05). The DBP main effects were qualified by a
significant Support Condition X Stressor Type interaction (F (2,
150) = 2.97, p = .05, 2 = .04; Figure 1). Follow-up mean compar-
isons revealed that, during the math task, those with their dog
present had lower DBP reactivity than those with their friend
present (p < .05). Participants who were alone had lower DBP
reactivity than those who had their dog or friend present (p’s <
.05). No significant effects were found for SBP.
    Cardiovascular recovery after stressors. Analyses of cardiovas-
cular responses during the recovery periods from the stressors
did not reveal significant main effects for support condition
or stressor type. However, there was a significant Support
Condition X Stressor Type interaction for SBP (F (2, 149) = 3.81,
p = .02, 2 = .05, Figure 2) and a trend interaction for HR (F (2,
149) = 2.42, p = .09, 2 = .03, Figure 2). Follow-up mean com-
parisons revealed that, during recovery from the math task,
those with their dog present tended to have lower SBP (p =
.08) and HR (p = .09) than those with their friend present. The
dog present condition did not differ from the alone condition.
Additionally, participants with their friend present tended to
have higher SBP (p = .08) and HR (p = .09) than those who were
alone.
    In summary, during the most stressful task (math), those
with their dog present had lower DBP reactivity than those
with a friend present. Additionally, across the stressor tasks,
participants with their dogs present had lower HR reactivity
than those with their friends present. Furthermore, during
the recovery period from the math task, there were trends for
participants with their dog present to have lower SBP and HR
than those with their friend present. These findings suggest
that, compared to friends, companion dogs are associated
with more cardiovascular benefits for their owners while ex-
periencing stress, as well as aiding in quicker cardiovascular
250			                          Journal of Sociology & Social Welfare
Figure 1. Cardiovascular Reactivity Means during the Math
and Cold Pressor Tasks.

              12

              10
                                                               *
              8
Mean Change

              6
                                                         *

              4                                      *
              2

              0
                   HR             SBP                    DBP
                         Math Task Reactivity
                        Alone      Dog      Friend

              7

              6

              5
Mean Change

              4

              3

              2

              1

              0
                   HR             SBP                    DBP
                        Cold Pressor Reactivity
                        Alone      Dog      Friend
*p < .05

recovery from stress. Next, for exploratory purposes we ex-
amined whether the significant findings of feeling evaluated
and threatened (evaluation apprehension) during the math
task explained the relationship between support condition and
DBPreactivity. That is, the presence of a close friend may have
Cardiovascular Benefits of Companion Dogs                        251
been more stressful than beneficial because of evaluation ap-
prehension concerns (Cacioppo & Petty, 1986; Guerin, 1989).

Figure 2. Cardiovascular Recovery Means after the Math and
Cold Pressor tasks.

              2.5

                2

              1.5

                1
Mean Change

              0.5
                                           †
                     †   †            †
                0

              -0.5

               -1

              -1.5       HR               SBP              DBP
                               Math Task Reactivity
               -2
                              Alone        Dog    Friend

              1.5

                1

              0.5
Mean Change

                0

              -0.5

               -1

                         HR               SBP              DBP
              -1.5
                              Cold Pressor Reactivity
                              Alone        Dog    Friend
†p < .09
252			                        Journal of Sociology & Social Welfare
Mediation Analysis
    For mediation analysis, we used the non-parametric boot-
strapping procedure of Preacher and Hayes (2008) for multi-
ple mediation. This approach tests the total indirect effect of
an independent variable (support condition) on a dependent
variable (cardiovascular reactivity) through a mediator or set
of mediators (feeling evaluated and threatened), and has the
capability to test the specific indirect effects if there are multi-
ple mediators in a set. Interpretation is accomplished by deter-
mining whether zero is contained within the 95% confidence
intervals. The statistical advantages of this method are that: (1)
multiple mediators can be tested simultaneously; (2) it does
not rely on a normal sampling distribution; (3) it can be used
with relatively small sample sizes; and (4) the number of infer-
ential tests is minimized, therefore reducing the likelihood of
Type 1 errors (Preacher & Hayes, 2008).
    We examined whether feeling threatened and evaluated,
as a set or individually, mediated the relationship between
support condition (dog, friend) and DBP reactivity to the math
task. This relationship was examined for DBP reactivity because
this was found to be a significant Support Condition X Stressor
Type interaction. The alone condition was excluded from these
analyses as they did not complete the evaluation apprehension
measure. The total and direct effects of support condition on
DBP reactivity during the math task were 3.59, p = .009 and
5.37, p = .001, respectively. The total indirect effect through
both mediators (feeling threatened and evaluated) was signifi-
cantly from zero (point estimate = -1.78, 95% BCa bootstrap
CI: -3.51 to -0.46). Therefore, as a set, feeling threatened and
feeling evaluated mediated the relationship between support
condition and DBP reactivity to the math task. Furthermore,
examination of the specific indirect effects indicated that only
feeling evaluated (a1b1 path) was a significant specific mediator
(point estimate = -1.79, 95% BCa bootstrap CI: -3.49 to -0.43). In
summary, these results suggest that participants with a friend
present had greater DBP reactivity to the math task than those
with a companion dog present because they felt more evalu-
ated by having their friend present.
Cardiovascular Benefits of Companion Dogs                   253

                           Discussion
    The main purpose of this study was to examine whether
having one’s companion dog present during and after stress
posed similar cardiovascular benefits as having a close friend
present, even when the relationship quality for both the dog
and friend was similarly positive. Our results indicated that
the presence of a companion dog was associated with lower
cardiovascular responses during stressors (i.e., DBP & HR)
and trends for lower cardiovascular responses during recov-
ery (i.e., SBP & HR) than having a friend present. The inter-
actions for reactivity (i.e., DBP) and recovery (i.e., SBP & HR
trends) revealed this was particularly true during the math
task. Moreover, mediation tests indicated that participants
with a friend present may have had higher cardiovascular re-
activity (i.e., DBP) during the math task because of evaluation
apprehension concerns, a finding that has been demonstrat-
ed in other research (Cacioppo & Petty, 1986; Guerin, 1989).
Unexpectedly, the alone participants’ responses were general-
ly comparable to those with a dog present (i.e., except for DBP
math reactivity). Guided by the stress buffering literature, we
had expected the participants in the alone condition to have the
highest levels of reactivity. An explanation may be that those
who were alone did not feel distressed because just knowing
that their dog was nearby (i.e., room or outside), without the
presence of a potentially evaluating friend, made these partici-
pants have similar cardiovascular responses as those with the
dog present in the room.
    In examining emotional responses, we found a trend for
participants with their dogs present to have a larger decrease
in unpleasant-activation emotions (i.e., feeling distressed,
nervous, and jittery) during recovery than those with a
friend present and significantly larger decrease than partici-
pants who were alone. Additionally, they felt less evaluated
and threatened during the math task than those with a friend
present. However, there were no differences in other emotions,
such as higher positive emotions in the dog condition com-
pared to the other conditions. This is surprising considering
that many owners claim to have intense emotional bonds with
their companion dogs (Cohen, 2002; Collis & McNicholas,
1998). Similarly, interactions with their animals make people
254			                        Journal of Sociology & Social Welfare
happier, less lonely, more relaxed, secure, and affectionate
(Barba, 1995). We know of no other experimental studies that
have examined emotional processes associated with the pres-
ence of one’s companion animal during stress. However, the
lack of results is consistent with social support research that
has been unable to find that emotional states differed as a func-
tion of support conditions (Gerin, Pieper, Levy, & Pickering,
1992; Lepore, Allen, & Evans, 1993). The emotional measure
we used in this study may not have been appropriate for cap-
turing the positive emotions that are associated with the bond
between an individual and his/her companion animal. Future
studies will need to explore how best to capture this emotional
experience with one’s companion animal.
     Uniquely, this study simultaneously examined individu-
als’ relationship quality with their companion animal and a
close friend. As expected, participants reported high levels
of bonding with their dogs, and both their friends and dogs
were helpful and important when needing emotional support.
Interestingly, the ratings for importance of one’s dog during
emotional support were significantly higher than the ratings
for friends. This may be due to these relationships fulfill-
ing different aspects of emotional support. Likewise, Bonas,
McNicholas, and Collis (2000) found that companion animals
(i.e., mainly dogs) scored higher than humans for fulfilling al-
liance, nurturance, and companionship needs. Human–animal
relationships may be characterized by less variability than
human friendships (i.e., friends may become too busy in their
own lives to be supportive) and many individuals report their
companion animals to be sources of unconditional love. We
also found that the positive aspects of relationship quality (i.e.,
SRI-importance, SRI-helpful) for participants’ dogs and friends
were not related to one another. This suggests that individuals
who value their relationships with their companion animals,
or perceive it as a highly positive one, may not be supplement-
ing for a lack of human support. In fact, companion animals
can help foster new human relationships as the presence of
one’s dog while walking or shopping may be associated with
increased interactions with strangers. Considering this, it may
be possible for individuals to have the best of both worlds: a
close bond with one’s companion animal and supportive close
friends.
Cardiovascular Benefits of Companion Dogs                     255
     A limitation of our study is that the participants were
healthy and young. These findings should be tested in older
adults and populations with special needs, such as chronically
ill or disabled individuals. Existing research has shown that
companion animals can help decrease loneliness and daily
stress as well as lower depression rates in such vulnerable pop-
ulations as those living with HIV/AIDS (Castelli et al., 2001),
with disabilities, and who are elders (Garrity, Stallones, Marx,
& Johnson, 1989; Steffens & Bergler, 1998). This is evidence that
companion animals are psychologically beneficial in such vul-
nerable populations; theoretically, such benefits should carry
over to effect physical health outcomes.
     In summary, our study indicates that even when relation-
ship quality is similarly positive for companion dogs and
close friends, dogs may provide more cardiovascular benefits
during stress and while recovering from stress. Particularly in
stressful circumstances where there is potential for evaluation,
having a friend present may actually make it more stressful.
Although it is not always possible to have one’s dog present
during stressors, these findings indicate that dogs are also as-
sociated with cardiovascular benefits afterwards. For example,
after a hard day at work, our companion animals may help
us rewind, both mentally and physically. Such findings also
have important implications for the psychological and physi-
cal well-being of a society that experiences significant levels of
stress. Stress can lead to the development and/or progression
of diseases (e.g., cardiovascular, infectious diseases, diabe-
tes, depression, etc.) by directly contributing to physiological
changes or, indirectly, by increasing health behaviors associ-
ated with disease risk and poor mental health outcomes. Our
relationships with our companion animals can reduce stress
levels and buffer the stress-related changes that occur in the
body (i.e., increased blood pressure, inflammation, & stress
hormones). Furthermore, they may encourage individuals to
engage in healthy behaviors to cope with stress, such as choos-
ing to walk the dog instead of attempting to cope through
more maladaptive behaviors (e.g., alcohol, drug use, overeat-
ing, etc.). Such benefits gained with our companion animal re-
lationships may be reflected on a society level through reduc-
tions in stress-related diseases (e.g., cardiovascular, infectious
diseases, diabetes, etc.) and improvements in psychological
256			                           Journal of Sociology & Social Welfare
well-being. Consequently, society holds a mutually beneficial
and rewarding relationship with companion animals—we
provide shelter and care for them and, in turn, the emotional
bonds we share help reduce our stress and the development of
stress-related diseases.

Acknowledgement: We would like to thank the participants, their
companion dogs and friends for participating in this study. This
study was supported by the National Institutes of Health under a
Ruth L. Kirschstein National Research Service Award (F31 HL081963-
01) from the National Heart, Lung, Blood Institute awarded to the
first author. Research Support: NIH NRSA F31 HL081963-01.
Dr. Campo is currently at the Program in Integrative Medicine, Dept
of Physical Medicine & Rehabilitation, University of North Carolina–
Chapel Hill.

                             References
Allen, K., Blascovich, J., & Mendes, W. B. (2002). Cardiovascular
    reactivity and the presence of pets, friends, and spouses: The
    truth about cats and dogs. Psychosomatic Medicine, 64(5), 727-739.
Allen, K., Shykoff, B. E., & Izzo, J. L., Jr. (2001). Pet ownership, but
    not ace inhibitor therapy, blunts home blood pressure responses
    to mental stress. Hypertension, 38(4), 815-820.
Barba, B. E. (1995). A critical review of research on the human/
    companion animal relationship: 1988 to 1993. Anthrozoos, 8, 9-19.
Barrett, L. F., & Russell, J. A. (1998). Independence and bipolarity
    in the structure of current affect. Journal of Personality and Social
    Psychology, 74, 967-984.
Blascovich, J., & Mendes, W. B. (Eds.). (2000). Challenge and threat
    appraisals: The role of affective cues. New York: Cambridge
    University Press.
Bonas, S., McNicholas, J., & Collis, G. M. (2000). Companion animals
    and us: Exploring the relationships between people and pets. New
    York: Cambridge University Press.
Cacioppo, J. T., & Petty, R. E. (1986). Social processes. New York:
    Guilford Press.
Campo, R. A., Uchino, B. N., Holt-Lunstad, J., Vaughn, A., Reblin,
    M., & Smith, T. W. (2009). The assessment of positivity and
    negativity in social support networks: The Social Relationships
    Index. Journal of Community Psychology, 37, 471-486.
Castelli, P., Hart, L. A., & Zasloff, R. L. (2001). Companion cats and
    the social support systems of men with AIDS. Psychological
    Reports, 89(1), 177-187.
Cardiovascular Benefits of Companion Dogs                             257
Cohen, S., & Wills, T. A. (1985). Stress, social support, and the
    buffering hypothesis. Psychological Bulletin, 98(2), 310-357.
Cohen, S. P. (2002). Can pets function as family members? Western
    Journal of Nursing Research, 24, 621-638.
Collis, G. M., & McNicholas, J. (1998). A theoretical basis for health
    benefits of pet ownership: Attachment versus psychological support.
    Thousand Oaks, CA: Sage Publications.
Crawford, E. K., Worsham, N. L., & Swinehart, E. R. (2006). Benefits
    derived from companion animals, and the use of the term
    “attachment.” Anthrozoos, 19, 98-112.
Dimitrijevic, I. (2009). Animal-assisted therapy—a new trend in the
    treatment of children and adults. Psychiatra Danubina, 21(2), 236-
    241.
Epstein, R. H., Huffnagle, S., & Bartkowski, R. R. (1991). Comparative
    accuracies of a finger blood pressure monitor and an oscillometric
    blood pressure monitor. Journal of Clinical Monitoring and
    Computing, 7(2), 161-167.
Feldman, P. J., Cohen, S., Hamrick, N., & Lepore, S. J. (2004).
    Psychological stress, appraisal, emotion, and cardiovascular
    response in a public speaking task. Psychology and Health, 19, 353-
    368.
Friedmann, E., & Thomas, S. A. (1995). Pet ownership, social support,
    and one-year survival after acute myocardial infarction in the
    Cardiac Arrhythmia Suppression Trial (CAST). American Journal
    of Cardiology, 76(17), 1213-1217. doi: S0002914999803439 [pii].
Garrity, T. F., Stallones, L., Marx, M. B., & Johnson, T. P. (1989). Pet
    ownership and attachment as supportive factors in the health of
    the elderly. Anthrozoos, 3, 35-44.
Gerin, W., Pieper, C., Levy, R., & Pickering, T. G. (1992). Social support
    in social interaction: A moderator of cardiovascular reactivity.
    Psychosomatic Medicine, 54(3), 324-336.
Guerin, B. (1989). Reducing evaluation effects in mere exposure. The
    Journal of Social Psychology, 129, 183-190.
Heinrichs, M., Baumgartner, T., Kirschbaum, C., & Ehlert, U. (2003).
    Social support and oxytocin interact to suppress cortisol and
    subjective responses to psychosocial stress. Biological Psychiatry,
    54(12), 1389-1398. doi: S0006322303004657 [pii].
Holt-Lunstad, J., Uchino, B. N., Smith, T. W., & Hicks, A. (2007). On the
    importance of relationship quality: The impact of ambivalence in
    friendships on cardiovascular functioning. Annals of Behavioral
    Medicine, 33(3), 278-290. doi: 10.1080/08836610701359795.
Krause-Parello, C. A. (2008). The mediating effect of pet attachment
    support between loneliness and general health in older females
    living in the community. Journal of Community Health Nursing,
    25(1), 1-14. doi: 10.1080/07370010701836286.
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, & coping. New
    York: Springer.
258			                            Journal of Sociology & Social Welfare
Lepore, S. J., Allen, K. A., & Evans, G. W. (1993). Social support lowers
    cardiovascular reactivity to an acute stressor. Psychosomatic
    Medicine, 55(6), 518-524.
Marcus, D. A., Bernstein, C. D., Constantin, J. M., Kunkel, F. A.,
    Breuer, P., & Hanlon, R. B. (2012). Animal-assisted therapy at an
    outpatient pain management clinic. Pain Medicine, 13(1), 45-57.
    doi: 10.1111/j.1526-4637.2011.01294.x.
McConnell, A. R., Brown, C. M., Shoda, T. M., Stayton, L. E., & Martin,
    C. E. (2011). Friends with benefits: on the positive consequences
    of pet ownership. Journal of Personality and Social Psychology,
    101(6), 1239-1252. doi: 10.1037/a0024506.
Nagaswa, M., Mogi, K., & Kikusui, T. (2009). Attachment between
    humans and dogs. Japanese Psychological Research, 51, 209-221.
Odendaal, J. S. (2000). Animal-assisted therapy—magic or medicine?
    Journal of Psychosomatic Research, 49(4), 275-280. doi: S0022-
    3999(00)00183-5 [pii].
Pierce, G. R., Sarason, I. G., Sarason, B. R., Solky-Butzel, J. A., & Nagle,
    L. C. (1997). Assessing the quality of personal relationships.
    Journal of Social and Personal Relationships, 14, 339-356.
Player, M. S., King, D. E., Mainous, A. G., III, & Geesey, M. E. (2007).
    Psychosocial factors and progression from prehypertension to
    hypertension or coronary heart disease. Annals of Family Medicine,
    5(5), 403-411. doi: 10.1370/afm.738.
Poresky, R. H., Hendrix, C., Mosier, J. E., & Samuelson, M. L. (1987).
    The Companion Animal Bonding Scale: Internal reliability and
    construct validity. Psychological Reports, 60, 743-746.
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling
    strategies for assessing and comparing indirect effects in multiple
    mediator models. Behavioral Research Methods, 40(3), 879-891.
Rosengren, A., Hawken, S., Ounpuu, S., Sliwa, K., Zubaid, M.,
    Almahmeed, W. A., Blackett, K. N., Sitthi-amorn, C., Sato, H., &
    Yusuf, S. (2004). Association of psychosocial risk factors with risk
    of acute myocardial infarction in 11119 cases and 13648 controls
    from 52 countries (the INTERHEART study): Case-control study.
    Lancet, 364(9438), 953-962. doi: 10.1016/S0140-6736(04)17019-0.
Serpell, J. (1991). Beneficial effects of pet ownership on some aspects
    of human health and behaviour. Journal of the Royal Society of
    Medicine, 84(12), 717-720.
Sherwood, A. (1993). Use of impedance cardiography in cardiovascular
    reactivity research. Washington, DC: American Psychological
    Association.
Steffens, M. C., & Bergler, R. (1998). Blind people and their dogs.
    Thousand Oaks, CA: Sage.
Templer, D. I., Salter, C. A., Dickey, S., Baldwin, R., & Veleber, D.
    (1981). The construction of a Pet Attitude Scale. Psychological
    Record(31), 343-348.
Cardiovascular Benefits of Companion Dogs                            259
Treiber, F. A., Kamarck, T., Schneiderman, N., Sheffield, D., Kapuku,
    G., & Taylor, T. (2003). Cardiovascular reactivity and development
    of preclinical and clinical disease states. Psychosomatic Medicine,
    65(1), 46-62.
Uchino, B. N., Cacioppo, J. T., & Kiecolt-Glaser, J. K. (1996). The
    relationship between social support and physiological processes:
    A review with emphasis on underlying mechanisms and
    implications for health. Psychological Bulletin, 119(3), 488-531.
Uchino, B. N., Holt-Lunstad, J., Uno, D., & Flinders, J. B. (2001).
    Heterogeneity in the social networks of young and older adults:
    Prediction of mental health and cardiovascular reactivity during
    acute stress. Journal of Behavioral Medicine, 24(4), 361-382.
Uno, D., Uchino, B. N., & Smith, T. W. (2002). Relationship quality
    moderates the effect of social support given by close friends
    on cardiovascular reactivity in women. International Journal of
    Behavioral Medicine, 9(3), 243-262.
Walsh, F. (2009). Human-animal bonds II: The role of pets in family
    systems and family therapy. Family Process, 48(4), 481-499. doi:
    10.1111/j.1545-5300.2009.01297.x.
Winefield, H. R., Black, A., & Chur-Hansen, A. (2008). Health effects
    of ownership of and attachment to companion animals in an
    older population. International Journal of Behavioral Medcine, 15(4),
    303-310. doi: 10.1080/10705500802365532.
You can also read